Private AI
Useful AI without a public model data path.
Private AI in BlackBox Node means local retrieval, local model execution, controlled source access, and permission-filtered context for organisations that cannot responsibly send sensitive data to public cloud AI services.
Data path
The sensitive path stays local
The product direction avoids sending confidential documents, prompts, retrieved context, or generated sensitive answers to external cloud model providers.
Runtime abstraction
Local model runtime first
Ollama is the first runtime abstraction direction, with llama.cpp planned later for tighter packaging and control where needed.
Governed assistance
Assistive, not autonomous
The product should help professionals search, draft, compare, and review information while preserving professional judgement and local governance.
Private AI path
Local context stays under local control.
The private AI story joins local storage, local retrieval, local generation, and professional review in one controlled appliance pattern.
-
01
Keep data local
Sensitive content remains in the client-controlled appliance environment.
-
02
Select permitted context
Retrieval is filtered by role, user, matter, client, and case permissions.
-
03
Run local model
A local model runtime receives only permitted context for answer support.
-
04
Review with citations
Professionals review generated drafts against source references.
-
05
Audit usage
The appliance records non-sensitive operational activity locally.
Public website boundary
This site explains the product. It is not the appliance.
This public website is product information only. Do not submit confidential client, patient, case, investigation, regulated, or commercially sensitive data through the public website.